Using Word Embeddings in Detection of Temporal Expressions in Turkish Texts
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Developing systems for automatically detection of date, time, duration and set expressions containing time information in texts is within the scope of Natural Language Processing research field. When studies for Turkish in the literature are reviewed, it is observed that only date and time expressions are included in the expressions detected by the models developed within the scope of Named Entity Recognition. There are studies to develop only rule-based systems on the subject of detection of temporal expressions in Turkish. Within the scope of this study, first Artificial Neural Networks based model for the detection of temporal expressions in Turkish texts is developed. The input of the developed model is word embeddings. In this study, the developed model success with using word embeddings built by different methods is measured on a dataset consisting of Turkish complaint texts collected from internet websites. By comparing the success of word embeddings on the detection of temporal expressions with the coverage percentages of word embeddings on the dataset, it is concluded that there is no correlation between them.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it